Paper: arXiv 2401.12652

Abstract

In this paper, we present ECL, a novel multi-modal dataset containing the textual and numerical data from corporate 10K filings and associated binary bankruptcy labels. Furthermore, we develop and critically evaluate several classical and neural bankruptcy prediction models using this dataset. Our findings suggest that the information contained in each data modality is complementary for bankruptcy prediction. We also see that the binary bankruptcy prediction target does not enable our models to distinguish next year bankruptcy from an unhealthy financial situation resulting in bankruptcy in later years. Finally, we explore the use of LLMs in the context of our task. We show how GPT-based models can be used to extract meaningful summaries from the textual data but zero-shot bankruptcy prediction results are poor. All resources required to access and update the dataset or replicate our experiments are available on github.com/henriarnoUG/ECL.

Complexity vs Empirical Score

  • Math Complexity: 3.0/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper focuses on dataset construction and empirical model evaluation (classical ML, neural networks, LLMs) with code and data availability, indicating practical implementation. The mathematics involved is standard for machine learning and statistics, without advanced theoretical derivations.

Research Flowchart

  flowchart TD
  A["Research Goal: Develop Multi-Modal Dataset & Models for Bankruptcy Prediction"] --> B["Dataset Creation"]
  B --> B1["Text: 10K Filings"]
  B --> B2["Numeric: Financial Ratios"]
  B --> B3["Labels: Binary Bankruptcy"]
  B --> C["Model Training & Evaluation"]
  C --> C1["Classical Models"]
  C --> C2["Neural Models"]
  C --> C3["LLMs / GPT Analysis"]
  C --> D["Key Findings & Outcomes"]
  D --> D1["Modality Complementarity"]
  D --> D2["Temporal Limitations of Binary Target"]
  D --> D3["LLM Summarization vs. Zero-shot Prediction"]